Download Is It Reasonable to Treat All Calcified Stenotic Aortic Valves

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Coronary artery disease wikipedia , lookup

Management of acute coronary syndrome wikipedia , lookup

Lutembacher's syndrome wikipedia , lookup

Marfan syndrome wikipedia , lookup

Turner syndrome wikipedia , lookup

Hypertrophic cardiomyopathy wikipedia , lookup

Quantium Medical Cardiac Output wikipedia , lookup

Jatene procedure wikipedia , lookup

Mitral insufficiency wikipedia , lookup

History of invasive and interventional cardiology wikipedia , lookup

Aortic stenosis wikipedia , lookup

Transcript
Journal of the American College of Cardiology
© 2008 by the American College of Cardiology Foundation
Published by Elsevier Inc.
Vol. 51, No. 5, 2008
ISSN 0735-1097/08/$34.00
doi:10.1016/j.jacc.2007.10.023
Valvular Heart Disease
Is It Reasonable to Treat All Calcified
Stenotic Aortic Valves With a Valved Stent?
Results From a Human Anatomic Study in Adults
Rachid Zegdi, MD, PHD,*†‡ Vlad Ciobotaru, MD,*† Miléna Noghin, MD,† Ghassan Sleilaty, MD,†
Antoine Lafont, MD, PHD,‡§ Christian Latrémouille, MD, PHD,*† Alain Deloche, MD,*†
Jean-Noël Fabiani, MD*†‡
Paris, France
Objectives
This study was designed to study the behavior of a stent deployed inside human stenotic aortic valves.
Background
Endovascular valved stent (VS) implantation is a promising new therapy for patients with severe calcific aortic
stenosis (AS). The precise characteristics of stent deployment in humans have been poorly studied so far.
Methods
Thirty-five patients with severe AS were included in the study. Sixteen patients (46%) had bicuspid aortic valves.
A self-expandable stent specifically designed for VS implantation was deployed intraoperatively inside the aortic
valve before surgical aortic valve replacement.
Results
In tricuspid aortic valves, the shape of stent deployment was circular, triangular, or elliptic in 68%, 21%, or 11%, respectively. Noncircular stent deployment was frequent in bicuspid aortic valves (the elliptic deployment being the rule
[79%]), and stent underdeployment was constant. The incidence of gaps between the stent external surface and the
aortic valve did not differ between tricuspid and bicuspid valves (58% vs. 43%; p ⫽ 0.49). Sharp calcific excrescences
protruding inside the stent lumen were present in 3 cases (9%). Ex vivo study of a homemade VS confirmed that the regularity of the coaptation line of the leaflets was critically dependent on the presence or the absence of stent misdeployment.
Conclusions
Stent misdeployment was constant in bicuspid valves and occurred in one-third of cases of tricuspid valves. Premature failure of implanted VS (secondary to valve distortion or traumatic injury to the leaflets by calcific excrescences) might be an important concern in the future. (J Am Coll Cardiol 2008;51:579–84) © 2008 by the
American College of Cardiology Foundation
Aortic stenosis (AS) is the most frequent valvular disease in
Western countries, and its incidence is expected to increase
with population aging. Until recently, surgical aortic valve
replacement (AVR) was the only effective treatment for the
most severe forms, providing relief of symptoms and an
increase in life expectancy (1).
Recently, endovascular aortic valve implantation (AVI)
has been proved to be feasible in AS, using either self- or
balloon-expandable valved stents (VS) (2– 4). Endovascular
AVI was first performed in compassionate cases and then in
high-risk surgical patients. However, AVI is not AVR,
because the native aortic valve is preserved in the former and
From the *Université René Descartes, Paris, France; †AP-HP, Assistance Publiquehôpitaux de Paris, Service de Chirurgie Cardio-Vasculaire, Hôpital Européen
Georges Pompidou, Paris, France; ‡Inserm U849, Faculté de Necker, Paris, France;
and §AP-HP, Assistance Publique-hôpitaux de Paris, Service de Cardiologie, Hôpital
Européen Georges Pompidou, Paris, France. Dr. Zegdi is a stock owner of Cormove,
a company developing a percutaneous valve.
Manuscript received June 21, 2007; revised manuscript received September 18,
2007, accepted October 8, 2007.
removed in the latter. Preservation of the native valve during
the endovascular procedure may be responsible, per se, for
complications such as coronary obstruction or perivalvular
leakage (2– 4).
Recourse to percutaneous AVI is expected to increase in the
future. However, the actual clinical experience remains limited,
and many problems need to be solved. Aside from coronary
obstruction and perivalvular leakage, conservation of the native
aortic valve may lead to other, still unknown complications.
Whether their occurrence depends on the anatomic type of the
aortic valve, that is, bicuspid or tricuspid, is also unanswered.
In an attempt to elucidate these 2 questions, a human
anatomic study was undertaken in patients scheduled for
nonemergent surgical AVR.
Materials and Methods
During a 5-month period, 35 patients (15 men; median age
71.5 years, range 40 to 90 years) with severe AS and
scheduled for AVR were included in the study. Median
preoperative aortic valve area was 0.7 cm2 (range 0.3 to 1 cm2),
Zegdi et al.
Valved Stents in Aortic Stenosis
580
JACC Vol. 51, No. 5, 2008
February 5, 2008:579–84
and mean systolic transaortic
valve gradient (median) was 43
mm Hg (range 10 to 92 mm
AS ⴝ aortic stenosis
Hg). Main pre-operative cliniAVI ⴝ aortic valve
cal and echocardiographic data
implantation
are reported in Tables 1 and 2.
AVR ⴝ aortic valve
The study was approved by the
replacement
local institutional review board,
VS ⴝ valved stent
and all patients gave their informed consent.
A braided nitinol self-expandable stent from Laboratoires
Perouse (Ivry-le-Temple, France) was used for intraoperative aortic valve “stenting.” The stent was designed specifically for endovascular valve replacement (5). Its diameter at
rest was 26 mm and was calibrated to develop a radial force
of around 1 kg when deployed in a 20-mm diameter orifice.
The stent was compressed by traction on a suture placed
around it and then was manually loaded into a 5-ml syringe
with a cut distal extremity used for stent delivery.
During surgery, after cardioplegic arrest of the heart, the
ascending aorta was opened and the aortic valve exposed.
The distal extremity of the syringe was positioned inside the
exposed aortic valve. The stent was delivered under visual
control inside the aortic valve by slowly pushing the piston
of the syringe. Then, the final stent deployment was
carefully analyzed. The shape of the deployed stent was
noticed, and a special attention was given to the presence of
any gaps between the stent external surface and the leaflets’
inner surface. These gaps are likely to be the anatomical cause
of the periprosthetic leaks observed in patients after percutaneous AVI. The gaps were checked with an oblique hook the
same way perivalvular dehiscence after prosthetic valve replacement is looked for. Once the analysis was achieved, the stent
was retrieved with a surgical forceps, and the AVR was
performed. The aortic valve “stenting” lasted no more than
2 min.
To analyze the impact of stent deployment on valve
geometry, a homemade pericardial VS was constructed.
Abbreviations
and Acronyms
PostoperativeClinical
Preoperative
Outcome
Characteristics
of the Study Population
and
Preoperative Clinical Characteristics and
Table 1
Postoperative Outcome of the Study Population
Baseline Characteristics
Age (yrs)
Gender (M/F)
Systemic hypertension
Diabetes mellitus
Coronary artery disease
Renal failure requiring hemodialysis
n (%) or Median (Range)
71.5 (40–90)
15 (43)/20 (57)
20 (57)
5 (14)
14 (40)
1 (3)
Post-operative outcome
30-day mortality
1 (3)
Low cardiac output syndrome
3 (9)
Reoperation for bleeding
1 (3)
Myocardial infarction
Stroke
0
0
Reoperation for bleeding
1 (3)
Mediastinitis
1 (3)
Pre-Operative
Data
of the Study
Echocardiographic
Population
Pre-Operative Echocardiographic
Table 2
Data of the Study Population
n (%) or Median (Range)
Aortic valve area (cm2)
0.7 (0.3–1)
Mean aortic gradient (mm Hg)
43 (10–92)
Aortic annulus diameter (mm)
19 (17–22)
Ejection fraction (%)
65 (20–87)
Ejection fraction ⱕ30%
2 (6)
Left ventricular end-diastolic diameter (mm)
51 (36–65)
Septal thickness (mm)
12 (9–16)
This valve was calibrated for implantation inside a circular
orifice with a 20-mm diameter and was also implanted in a
triangular (equilateral) or an elliptic (0.73 eccentricity)
orifice with a circumference identical to that of the circular
orifice. Implantation was also performed in a 17-mm
circular orifice to mimic the behavior of an oversized VS.
The influence of annular calcifications located close to a VS
commissure was also studied; this was mimicked by the
presence of a bulging inside a circular orifice.
Data are expressed as median (range) for continuous variables and as percentage for categorical variables. Comparisons
between categorical variables were performed with the Fischer
exact test. A p value ⬍0.05 was considered significant.
Results
Bicuspidy was present in 16 patients (46%). Stent delivery
was impossible in 2 patients (6%) who had severely stenotic
bicuspid valves.
After deployment, the aortic cross section of the stent
could be basically categorized into 3 aspects that were
grossly circular, elliptic, or triangular (Fig. 1). The stent
deployment was highly dependent on valve anatomy
(Table 3). The circular deployment was observed in 15
patients (45%), significantly more frequently in tricuspid
than in bicuspid valves (68% [n ⫽ 13] vs. 14% [n ⫽ 2]; p ⫽
0.004). The triangular aspect, usually mild, was seen mainly
in tricuspid valves (n ⫽ 4; 21% of tricuspid valves). This
aspect reflected the calcified leaflets’ rigidity, preventing
them from being sufficiently bent and correctly applied
against the stent’s external surface. The elliptic aspect was
common in bicuspid valves (n ⫽ 11; 79% of bicuspid valves).
In tricuspid valves, the median deployed stent’s external
diameter was 19 mm (range 17 to 20 mm) for the 13 cases
with a circular stent deployment. The median aortic annulus
diameter (as determined from the preoperative transthoracic
echocardiographic measurement) was 18 mm (range 17 to
21 mm). Underdeployment (arbitrarily defined as a ⱖ3 mm
difference between the aortic annulus and the stent external
diameter) was absent.
In bicuspid valves, the external diameter of the circularly
deployed stent and of the aortic annulus were 18 and 21 mm
in case 1 and 17 and 21 mm in case 2, respectively.
Underdeployment was present in both cases.
Zegdi et al.
Valved Stents in Aortic Stenosis
JACC Vol. 51, No. 5, 2008
February 5, 2008:579–84
Figure 1
581
Different Shapes of Stent Deployment Encountered
Circular (A), triangular (B), and elliptic (C and D). Note the round calcifications crossing the stent frame.
A gap between the stent external surface and the inner
surface of the aortic valve was identified in 17 patients
(49%). Theses gaps were located exclusively at the level of
the commissures. Their incidence was not dependent on the
valve pathology (58% [n ⫽ 11] in tricuspid vs. 43% [n ⫽ 6]
in bicuspid valves; p ⫽ 0.49). However, the presence of a
periprosthetic gap depended on the shape of stent deployment (Fig. 2). The highest rate (100%) was observed with
the triangular shape and the lowest with the circular one
(33%). In tricuspid valves, the gaps involved 1, 2, or 3
commissures in 7 (37%), 4 (21%), and 0 cases, respectively.
The aortic leaflets’ ventricular surface was always irregular, and calcifications almost constantly crossed the stent
frame. These calcifications were usually small, round, and
smooth (Fig. 1). In some cases (n ⫽ 3; 9%), however, these
calcifications were particularly sharp, protruding into the
stent lumen (Fig. 3).
The anchorage of the present stent to the surrounding
tissue was judged firm by surgeons, and its removal was
impossible without partial recompression. Aortic valve replacement was performed with a bioprosthetic or a mechanical valve in 31 (89%) and 4 (11%) patients, respectively.
After implantation of the homemade VS inside a circular
20-mm orifice, the coaptation line of the leaflets was regular
and symmetric, reproducing the “Mercedes sign” (Fig. 4). In
all other circumstances (undersized circular orifice, elliptic
or triangular orifice, asymmetric circular orifice), the coaptation line was disorganized, reflecting leaflet distortion
(Figs. 4 and 5). The severity of valve distortion markedly
depended on the location of the commissures inside the
triangular orifice (Fig. 6).
Discussion
Endovascular AVI has recently been proved feasible in AS
using either self- or balloon-expandable VS (2– 4). This
promising new therapy might be beneficial to a great
number of patients at high surgical risk or even patients for
Stent Shape
According
to After
AorticDeployment
Valve Pathology
Stent Shape After Deployment
Table 3
According to Aortic Valve Pathology
Stent Shape
Tricuspidy (n ⴝ 19)
Bicuspidy (n ⴝ 14)
Circular, n (%)
13 (68)
2 (14)
Elliptic, n (%)
2 (11)
11 (79)
Triangular, n (%)
4 (21)
1 (7)
Figure 2
Incidence of Peri-Stent Gaps According to the Aortic
Valve Anatomy and the Shape of Stent Deployment
582
Figure 3
Zegdi et al.
Valved Stents in Aortic Stenosis
Sharp Calcific Excrescences Crossing the
Stent Frame, Protruding Inside the Aortic Lumen
whom surgery is denied. The few clinical series currently
available have pointed out several procedure-related complications, such as valve migration, coronary obstruction, or
perivalvular leak (2– 4).
Concerns have been raised regarding the potential traumatic injury to the pericardial leaflets of the current VS
Figure 4
JACC Vol. 51, No. 5, 2008
February 5, 2008:579–84
during the crimping or loading process into the delivery
catheter. For some models, VS deployment requires balloon
dilation. The friction of the pericardial leaflets against the stent
frame during deployment may also alter their structure or their
mechanical properties, with potential negative effects on longterm durability (3). Mid-term or long-term follow-up are still
lacking, hence the impact of these potentially deleterious
processes on VS durability remains unknown.
The results of the present study suggest that durability of
the VS implanted inside a stenotic aortic valve might also be
altered through other mechanisms. In 9% of the cases, sharp
calcific excrescences protruded inside the stent lumen.
Repeated contact with a pericardial (or a polymeric) leaflet
during the aperture-closure cycle of the valve may lead to
leaflet perforation and valve dysfunction. Such lesions have
been described by surgeons during bioprosthetic valve replacement when suture knots were left too long (6). Only
sufficiently covered VS might be protected from the potentially harmful effect of these excrescences when the calcific
aortic valve is left in place.
The major finding of this study was the high incidence of
stent misdeployment (100% in bicuspid valves; one-third in
tricuspid valves). For optimal functioning, VS requires an
adequately sized cylindrical deployment of the stent part
supporting the leaflets. Any stent misdeployment may lead
Influence of Size or Shape of the Orifice on the Valved Stent Deployment
No leaflet distortion was present with the valved stent (VS) deployed inside the circular orifice with a 25-mm diameter (A).
Conversely, distortion occurred after deployment of the VS in an elliptic (B), a triangular (C), or an undersized circular orifice (D).
Zegdi et al.
Valved Stents in Aortic Stenosis
JACC Vol. 51, No. 5, 2008
February 5, 2008:579–84
Figure 5
583
Leaflet Distortion in the Presence of Annular Calcification Close to One Commissure of the Deployed Valved Stent
Calcification, mimicked by an irregularity (left, arrow) inside the circular orifice, leading to valve distortion (right).
to valve distortion (Figs. 4 to 6). Valve distortion increases
stress on 1 or more leaflets, especially at the level of the
commissures, resulting in premature failure by leaflet tear or
fibrosis, as demonstrated in an animal study (7).
Valve distortion may also occur despite circular stent deployment (Fig. 4). Such a situation may theoretically be observed
with oversized VS implantation. Oversizing has been proposed
by others (4,8) to reduce the incidence of periprosthetic leak. If
the aortic valve and/or annulus do not strictly accommodate
the oversized VS, valve distortion may occur because of excess
of pericardial tissue relatively to the stent orifice area.
Calcific bicuspidy is a frequent condition in severe AS,
with a 50% incidence in surgical series (9). Patients with a
bicuspid aortic valve are usually younger with, therefore, a
better post-operative long-term survival. Stent misdeployment in bicuspid valves was 100% in this anatomic study.
Underdeployment was constant owing to the limited valve
opening of bicuspid aortic valves. Furthermore, noncircular
deployment of the stent was the rule (86%), increasing VS
Figure 6
distortion attributable to stent underdeployment. Based on
these data, a high rate of patient–VS mismatches and of VS
premature failure secondary to valve distortion would therefore be expected. Because of the negative effect of patient–
prosthesis mismatch on midterm survival (10), the current
procedural risk of endovascular VS implantation, and the
potentially increased perioperative risk of a subsequent surgical
AVR, we believe that endovascular VS implantation should
not be recommended in patients with calcific bicuspid AS.
Aortic valve regurgitation was the most prevalent valvular
complication reported following endovascular aortic VR (2– 4).
The AVR was paravalvular in origin, and its severity may be
partially reduced by balloon redilation in balloon-expandable
VS (4, 9). With self-expandable VS, their incidence might
decrease with time (3). The present anatomic study confirmed
the high incidence (58% in tricuspid valves) of peri-stent
dehiscence, which was comparable to the echocardiographic
incidence (52%) in the series of Grube et al. (3). Leaks were
always located at the level of commissures.
Valve Distortion Secondary to the Valved Stent Deployment Inside a Triangular Orifice
The severity of valve distortion was clearly dependent on positioning of the valved
stent commissures inside the orifice (less severe in the left panel than in the right panel).
584
Zegdi et al.
Valved Stents in Aortic Stenosis
Aortic leaflets present rough ventricular surfaces in calcific AS. Round calcifications are almost constant and
multiple on this side (Fig. 1). Despite their small size, they
may have 2 consequences. First, their protrusion through
the stent frame may provide a better VS anchorage. Subsequent stent migration would therefore be impossible without a spontaneous in vivo stent recompression, an eventuality rather unlikely to occur with self-expandable stents.
Second, round calcifications may favor the occurrence of
valve distortion (Fig. 5) or even periprosthetic leak. This
may happen when part of the stent frame lies on the top of
a rather large calcification, creating a gap between the
external stent surface and the leaflets’ internal surface.
Technically speaking, this phenomenon may occur with any
type of stent but is less likely with braided stents, whose
frame can better accommodate this situation (as reflected by
the local deformation of the stent frame in Fig. 1). These 2
potential consequences are partly speculative and require
further evaluation.
The present study relied on the use of a self-expandable
stent. In tricuspid aortic valve stenosis, its radial force was
sufficient to achieve an adequate circular deployment in
many cases (68%) and a firm anchorage in all cases. We do
not believe that the use of a stiffer stent would have resulted
in better expansion in bicuspid aortic valves. In these valves,
because of the calcifications, the excursion of at least 1
leaflet (usually the leaflet with the raphe) (Fig. 1) is severely
restricted, even with the surgeon’s forceps. This restriction
has also previously been shown in earlier studies demonstrating the poor results of valvuloplasty in bicuspid AS (11).
However, it is possible that a better expansion might be
observed in tricuspid valves (this is the subject of an ongoing
study), but the risk of coronary obstruction might increase
with such stiffer stents.
It is also possible that balloon valvuloplasty before aortic
stenting (as is the case in clinical practice before VS implantation) may reduce the prevalence of noncircular stent deployment by “fragmentation” of the leaflet’s calcifications. Fragmentation might improve the pliability of the aortic valve
leaflets and, therefore, favor the circular deployment of the
stent.
Rare postmortem cases have been discussed (2,4), but no
systematic anatomic studies from cadaveric or autopsy series
and no radiographic studies of VS deployment have been
published so far. Such studies are required to corroborate or
invalidate the present data, especially if the procedure is
intended to be applied to a larger group of patients.
In conclusion, this first human anatomical study of
intraoperative aortic valve stenting in patients with AS
suggested that percutaneously implanted VS might show
premature failure through 2 previously unrecognized mechanisms: injury to the leaflets by calcific excrescences and
JACC Vol. 51, No. 5, 2008
February 5, 2008:579–84
valve distortion secondary to stent misdeployment. Owing
to the relatively high incidence of these unrecognized
mechanisms, physicians should wait for the mid- and
long-term results of the procedure in high-risk patients
before extending the indication to less risky surgical patients. The present study also has shown that stent misdeployment (and consequently valve distortion) was constant in
bicuspidy, suggesting that percutaneous AVI may have less
favorable results in bicuspid than in tricuspid AS.
Acknowledgments
The authors are deeply grateful to Witold Styrc (Flashmed,
Echternach, Luxembourg) and Ming Shen (Laboratoires Perouse, Ivry-le-Temple, France) for their technical assistance.
Reprint requests and correspondence: Dr. Rachid Zegdi, Hôpital Européen Georges Pompidou, Service de Chirurgie Cardiovasculaire, 20 rue Leblanc, 75908 Paris, France. E-mail: rzegdi@
hotmail.com.
REFERENCES
1. Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006
guidelines for the management of patients with valvular heart disease:
a report of the American College of Cardiology/American Heart
Association Task Force on Practice Guidelines (Writing Committee
to Revise the 1998 Guidelines for the Management of Patients With
Valvular Heart Disease). J Am Coll Cardiol 2006;48:e1–148.
2. Cribier A, Eltchaninoff H, Tron C, et al. Early experience with
percutaneous transcatheter implantation of heart valve prosthesis for
the treatment of end-stage inoperable patients with calcific aortic
stenosis. J Am Coll Cardiol 2004;43:698 –703.
3. Grube E, Laborde JC, Gerckens U, et al. Percutaneous implantation
of the Corevalve self-expanding valve prosthesis in high-risk patients
with aortic valve disease. The Siegburg first-in-man study. Circulation
2006;114:1616 –24.
4. Webb JG, Chandavimol M, Thompson CR, et al. Percutaneous aortic
valve implantation retrograde from the femoral artery. Circulation
2006;113:842–50.
5. Van Nooten G, Ozaki S, Herijgers P, Segers P, Verdonck P, Flameng
W. Distortion of the stentless porcine valve induces accelerated leaflet
fibrosis and calcification in juvenile sheep. J Heart Valve Dis 1999;8:
34 – 41.
6. Zegdi R, Khabbaz Z, Borenstein N, Fabiani JN. A repositionable
valved stent for endovascular treatment of deteriorated bioprostheses.
J Am Coll Cardiol 2006;48:1365– 8.
7. Garcia-Valentin A, Castellà M, Josa M, Mulet R. Bioprosthetic leaflet
perforation due to repetitive trauma by overknotted sutures. Eur
J Cardiothorac Surg 2006;29:106.
8. Walther T, Falk V, Borger MA, et al. Minimally invasive transapical
beating heart aortic valve implantation—proof of concept. Eur J Cardiothorac Surg 2007;31:9 –15.
9. Roberts WC, Ko JM. Frequency by decades of unicuspid, bicuspid,
and tricuspid aortic valves in adults having isolated aortic valve
replacement for aortic stenosis, with or without associated aortic
regurgitation. Circulation 2005;111:920 –5.
10. Tasca G, Mahgna Z, Perotti S, et al. Impact of prosthesis-patient
mismatch on cardiac events and midterm mortality after aortic valve
replacement in patients with pure aortic stenosis. Circulation 2006;
113:570 – 6.
11. Isner JM, Salem DN, Desnoyer MR, et al. Treatment of calcific aortic
stenosis by balloon valvuloplasty. Am J Cardiol 1987;59:313–7.